This article was downloaded by: [University of North Carolina]On: 08 October 2014, At: 14:28Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Atlantic Journal of CommunicationPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/hajc20

Data, Doppler, or Depth of Knowledge:How Do Television Stations DifferentiateLocal Weather?George L. Daniels a & Ginger Miller Loggins ba Department of Journalism , University of Alabamab A. Q. Miller School of Journalism and Mass Communications, KansasSate UniversityPublished online: 21 Jan 2010.

To cite this article: George L. Daniels & Ginger Miller Loggins (2010) Data, Doppler, or Depthof Knowledge: How Do Television Stations Differentiate Local Weather?, Atlantic Journal ofCommunication, 18:1, 22-35, DOI: 10.1080/15456870903340472

To link to this article: http://dx.doi.org/10.1080/15456870903340472

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Atlantic Journal of Communication, 18:22–35, 2010

Copyright © Taylor & Francis Group, LLC

ISSN: 1545-6870 print/1545-6889 online

DOI: 10.1080/15456870903340472

Data, Doppler, or Depth of Knowledge:How Do Television Stations Differentiate

Local Weather?

George L. DanielsDepartment of Journalism

University of Alabama

Ginger Miller LogginsA. Q. Miller School of Journalism and Mass Communications

Kansas Sate University

Although weather coverage has never received extensive research, previous studies have investigated

the importance of weather to viewers and to television station branding. In addition, such individual

features of weather reports like accuracy have been investigated, but no researcher has apparently

examined the overall content of day-to-day local televised weathercasts. This content analysis of

weathercasts in five medium to large Southern U.S. markets sheds light on the ways stations

differentiate themselves through their weathercasts’ content. It found that so many stations use

radar or claims of accuracy to differentiate themselves that the techniques do not result in product

differentiation. Other methods, by themselves or combined with accuracy and radar, may better

differentiate weathercasts.

With more than 21 named storms, the 2005 Hurricane Season ended as one of the most active

in recent history and devastated many areas in the Gulf Coast. Since 2005, fewer hurricanes

have made landfall, but deadly tornadoes have left their mark on Enterprise, Alabama, andGreensburg, Kansas; blizzards shut down Denver, Colorado; and floods crippled parts of

northeast Ohio. A decade ago, researchers found that weather was the number one news draw

in 11 of the nation’s top 20 television markets (Bowser, 1997).

The importance of televised weather was demonstrated in 2006 as local broadcast meteorol-

ogists in the Evansville, Indiana, and Des Moines, Iowa, markets found themselves at the centerof deadly weather events. In Iowa, twisters struck during the daytime and forced thousands to

evacuate a football stadium, but officials reported only one fatality (Higgins, 2005). Residents

Correspondence concerning this article should be addressed to George L. Daniels, Department of Journalism,

University of Alabama, Box 870172, Tuscaloosa, AL 35487-0172. E-mail: gdaniels@ua.edu

22

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

DIFFERENTIATING WEATHER 23

in two southern Indiana counties were not so fortunate when tornadoes came between 2 and

3 a.m. (Neal & Hupp, 2005), the second most likely time of day for tornadoes (Schneider,2005). Most people had no warning at that time of the morning, allowing the twisters to claim

the lives of 23 people and leave more than 150 injured. Such weather events make it easy to

understand why a television station’s commitment to local weather is vital for a community.

Beyond the recent incidents of severe weather, at least three other developments in the deliv-

ery of local weather information suggest the topic of local weather is worthy of investigation.First, some television viewers now get forecasts on their computer or via portable devices

such as cell phones before any newscasts air. A second recent change is the effort by some

companies to more efficiently provide local news weather segments by using meteorologists

at a centralized weather facility. Third, in the last few years, the American Meteorological

Society (AMS) upgraded its requirements to a new Certified Broadcast Meteorologist seal,

which unlike the traditional AMS seal specifically requires a degree in meteorology (Walker,2006). These developments demonstrate the constant change in the market, which along with

the continual fragmentation of audiences suggests a need for research on the role of weather

branding in station differentiation.

This study specifically focused on weather segments from stations in five large and medium-

sized media markets. The segments were analyzed on three dimensions: standard weather fare,tools and technology, and brand image. Given the prominence of the weather segment in a

local news operation’s image, this research considered the weather information delivered on a

day-to-day basis as a competitive product. By looking at weather as product, the analysis drew

on a conceptual framework from marketing and competitive strategy literature. Scholarly and

trade literature on weather forecasting and findings from previous media studies of brandingand product differentiation also influences the analysis.

LITERATURE REVIEW

Researching Weather

The literature provides ample evidence that local weather coverage has importance for viewersand local stations. For viewers, it is not only important, it is also remembered well. For instance,

one study found that the public preferred weather second only to the local economy in news

categories (Roberts & Dickson, 1984). A separate investigation showed that weather coverage

is the number one reason that people watch local news (Bowser, 1997), and a third study

conducted by a media consulting firm found that viewers rank weather as the most important

item in a newscast (Galetto, 1997). In addition, Neuman (1976) found that viewers recalledweather better than almost all other news story categories.

Newspaper articles also show that local television coverage of weather gains interest from

other media. In Indianapolis, newspapers felt the need to cover the “ratings wars with radar”

between the leading television stations (Schoettle, 2005). Likewise, the St. Petersburg Times

profiled the highly competitive lead broadcast meteorologists in the Tampa–St. Petersburgmarket based on their respective weather forecasting tools, bragging rights on superiority of

equipment and meteorological training (Colavecchio-Van Sickler, 2004).

As for weather’s importance to stations, such content takes almost 10% of a newscast’s time,

or 4 12

min, according to an early study of newscasts in the San Diego market (Wulfemeyer,

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

24 DANIELS AND LOGGINS

1982). This is based on the 1-hr newscast rather than the more typical half-hour local television

newscast today. A former Radio-Television News Director Association (RTNDA) chairman alsoemphasized weather’s importance for a station’s image: “A successful weather image is built

upon far more than the occasional weather disaster: : : : Stations cultivate their image through

countless days of routine forecasts and benign weather maps, punctuated by the occasional

big event” (Salsberg, 2003, p. 14). In addition, weather gets attention from the RTNDA,

which focused on local forecast credentials when it considered why viewers watch local news(O’Malley, 1999).

Critical scholars have also emphasized the importance of weather-related news in our society.

Sturken (2001) suggested that it can alleviate postmodern anxieties about such things as

fragmentation, rapid social change, and lack of meaning. Meister (2001) called TV meteorology

“priestly” in its speech of scientific culture and argued that the Weather Channel created

“weathertainment” that predicates and encourages consumer practices.Among media audiences, television stations appear to get most of the attention and are

working hard to retain their viewers through online media. In examining which public media

audiences used most often for weather information, Tan (1976) provided one of the first

journalism/mass communication studies focused on local weathercasts. Among the respondents

surveyed in 1973, 41% preferred television, followed closely by 40% for radio (Tan, 1976).In more recent research, when Hurricane Danny struck the Alabama–West Florida coast in

1997, a study showed the public relied on local television coverage and local radio reports

as their major information/news source (Piotrowski & Armstrong, 1998). The same study

showed that the Internet is gaining in prominence. A sizeable minority of the 325 respondents

noted using Internet weather sites and weather band radio for information on the hurricane.However, many television stations are moving into that media market, and a content analysis

of 128 Internet pages just 3 years ago showed television home pages offered significantly more

weather presentation tools than news radio and newspapers (Randle & Mordock, 2002). Even

as early as 1997 a strong Web presence paid off for one Arkansas station during tornado season.

One weekend when multiple tornadoes hit Little Rock, KTHV-TV’s site received 100,000 hits

(Galetto, 1997).For the television viewers, certain weather features appear more important than others.

Scholars have determined that when it comes to news preferences in the area of weather,

respondents ranked information about local conditions of highest interest followed by lo-

cal forecasts, national conditions, and national forecasts (Wulfemeyer, 1983). A more recent

dissertation-length study of viewer preferences found “new” weather graphics can improveviewer comprehension if the meteorologist takes time to explain the graphics (Siebert, 2000).

Accuracy is a growing concern for broadcasters. It was the focus of Broadcasting & Cable

magazine’s May 2007 cover story. The story examined concerns within the meteorological

community that in the highly competitive weather forecasting arena, local stations are risking

their credibility by providing longer range (i.e., 10-day) forecasts when conditions are virtuallyimpossible to predict (Malone, 2007). When WeatheRate became the first outside agency to

verify a station’s forecasting accuracy in 2004, it quickly found stations interested in touting the

company’s seal of approval (Potter, 2004). When it comes to weathercasters’ expertise, results

from a survey of 136 weathercasters showed that 56% had meteorology or related degrees

(Earl & Pasternack, 1991), and Walker (2006) highlighted the importance of a meteorological

degree and certification for weathercasters in medium-sized markets in his research.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

DIFFERENTIATING WEATHER 25

In addition to accuracy, ability and technology appear to be weather features of growing

importance. One of the first studies of journalism and mass communication programs’role in preparing broadcast meteorologists found nearly 8 in 10 professional broadcasters

thought weather should be covered more in the mass communication curriculum, whereas

only 48% of educators agreed (Davie, Auter, & Dinu, 2006). As for technology, the

RTNDA Communicator regularly features weather technologies that can boost prediction

accuracy and precision in its “New Products” columns and feature stories (Murrie, 2000a,2002).

But exactly what makes up a station’s normal weather segment? This study attempts to

answer that question even though past studies have had some success in finding the answer.

It appears that most of the time stations do get the broad conditions right. Gantz (1982)

found that almost all of the forecasts he verified (88%) were accurate in their rain/no rain

predictions. Forecasts were most frequently accurate for “next day” predictions, but only 5%of temperature forecasts were exactly right. In addition, the predictions made in the early

evening newscasts are generally just as accurate or inaccurate as the late evening newscasts

as the forecasts rarely change between two evening shows. As for basic content, Earl and

Pasternack (1991) found that most stations responding to their survey presented general weather

forecasts. Specialized forecasts were relatively uncommon whereas current temperature, winddirection, relative humidity, and wind speed were the most common topics in the general

weather segments. In her study of media coverage of a blizzard, Wilkins (1984) found that

the media generally focused on the event itself and provided relatively few analytic stories,

whereas Wulfemeyer (1982) examined style and found that the weathercaster often took on the

role of the “lovable buffoon.”

Branding

It has been said that consumers lack the motivation, capacity, or opportunity to process the

product information that they see and hear in a thoughtful or deliberative manner (McDowell,

2006). Thus, advertising and marketing researchers have suggested that strong brands assist

in this process by making the task of choosing a particular product much easier (McDowell,2006). By definition, a brand is a name, term, sign, design, or a unifying combination of those

elements that distinguish a product or service from its competitors.

Literature on branding emphasizes its importance. A study of 84 television general managers

found that most agree branding is a useful business tool and that a brand image will help

their station stand out (Chan-Olmsted & Kim, 2001). They also perceived “advertising” and

“community events” to be the most effective tactics for branding. In the minds of someresearchers, a station’s visual or auditory brand is the “hottest” of contemporary marketing

strategies (Bellamy & Traudt, 2000).

Weather seems to be an integral part of station branding. Those who have surveyed local

news directors have suggested the nightly focus on weather is crucial so that viewers know

where to look or listen when the weather takes a turn (Papper, 2002). In fact, some consultantshave suggested that weather image is as important as the news image in building viewers’

perception of a station. This idea of weather branding simply places a high-interest item of the

newscast at the top of a television station’s news promotional strategy. To facilitate this type of

branding, weather data providers such as the Weather Channel, the now-defunct NBC Weather

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

26 DANIELS AND LOGGINS

Plus,1 and AccuWeather have expanded their digital offerings so that the brand associated with

the weather data could be extended beyond a scheduled newscast to such things as a separatecable channel with continuous local weather information (Dickson, 2006).

Production Differentiation

A number of scholars have used product differentiation to examine local television news content.An analysis of early evening newscasts in three Midwestern markets showed half the stories

in Detroit, Michigan; Toledo, Ohio; and East Lansing, Michigan, were unique. The larger the

market, the more unique the stories it had (Atwater, 1984). A more recent product differentiation

study compared content of broadcast network news and all-news cable networks (Bae, 2000).

The three cable networks had significantly more unique stories than the broadcast networks.A separate study of the cable content suggested product differentiation allowed the 24-hr ca-

ble news channels to increase what’s known as horizontal diversity (Bae, 1999). The horizontal

diversity or horizontal production differentiation is a concept that comes from the management

strategy literature. It focuses on the differences in attribute variety among competing brands

(p. 64). For instance, two stations in the same market could diversify through sports opens.

One could feature audio sounders to signify the beginning of sports and another station mayuse a more subtle anchor toss straight to the sportscaster.

In his classic strategy text, Porter (1980) identified differentiation as one of three potentially

successful generic strategic approaches to outperforming firms in an industry. By differentiating

one’s product or service, the firm creates something that is perceived as being unique. The

resulting differentiation provides insulation against competitive rivalry, increases margins andachieves customer loyalty (Porter, 1980).

In laying out the models for product differentiation, Ireland (1987) acknowledged that the

key question is whether horizontal or vertical differentiation dominates. According to her

model, if there is horizontal dominance (some consumers like some products, whereas others

like another), one would expect little competition. Vertical dominance, on the other hand,

encourages intense competition among lower quality products. Furthermore, she suggested thatlow homogenous quality becomes the norm when horizontal product differentiation dominates,

whereas high-quality products occur when vertical dominance is possible (Ireland, 1987).

HYPOTHESES AND RESEARCH QUESTIONS

The product differentiation conceptual framework from the marketing and strategy literature

is helpful in examining how local stations position their weathercast in the highly competitive

arena of local television news. As has been reported by multiple newspapers (Colavecchio-VanSickler, 2004; Schoettle, 2005), television stations in many markets often position themselves

in the eyes of viewers by their radar in so-called ratings wars with radar. Thus, in what Ireland

(1987) conceptualized as vertical product differentiation, one would expect radar to be the

1NBC Universal purchased the Weather Channel from Landmark Communications in 2008 for a reported

$3.5 billion. NBC Weather Plus, a 4-year digital joint venture between NBC and its affiliated local stations, was

subsequently phased out.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

DIFFERENTIATING WEATHER 27

particular characteristic that stations hope consumers perceived as making a weather product

better. In this study, the investigators hypothesized that stations will differentiate weathercastsby radar more than other tools or technology (H1).

Because those in local broadcast news have suggested that successful weather image relies

on countless days of routine forecasts (Salsberg, 2003) and station managers have apparently

embraced branding as a useful business tool to help a station stand out (Chan-Olmsted & Kim,

2001), it was hypothesized that the most successful stations in a local market (the number oneor number two rated television stations) would be more likely than other television stations to

use a brand identity to differentiate their weathercasts (H2).

For years, local stations have used the abilities of their weather forecasters to accurately

predict the weather as a promotional strategy. Given the impact of forecast accuracy on their

promotion, it is likely that stations that use a degree guarantee as a product differentiation

strategy would have a higher degree of accuracy in predicting the next day’s high temperaturethan stations without the guarantee (H3). At the same time, Gantz’s (1982) research with

Indiana stations showed that only 5% of temperature forecasts were exactly right. In his study,

forecasts were most frequently accurate for “next day” predictions, but only 5% of temperature

forecasts were exactly right (Gantz, 1982). In this study, researchers also looked at how accurate

local meteorologists were in predicting the high temperatures in the next day’s forecast (RQ1).Beyond the promotional tools such as the degree guarantees and technologies like radar,

there are certain aspects of a local weather forecast that are the same regardless of station.

In the study, the investigators looked to identify what specific elements constituted so-called

standard weather fare (RQ2).

Over the last decade, the television industry has been characterized by consolidation asstation groups acquired more stations to increase their buyer power for television programming

and their seller power with advertisers. For example, Liberty Corporation (Greenville, South

Carolina) was acquired by Raycom Media (Montgomery, Alabama) in a $1 billion purchase

(Werner, 2005). Station groups typically employ similar strategies across local station prop-

erties. Thus, this study sought to determine the role television group ownership plays in a

station’s weathercast strategy (RQ3).

METHODOLOGY

Sample of Programs

The hypotheses were tested and research questions answered using a content analysis of 146evening weathercasts in five markets (Tampa-St. Petersburg, Florida; Birmingham, Alabama;

Oklahoma City, Oklahoma; Richmond, Virginia; and Columbia, South Carolina) from fall 2003

to winter 2005. The convenience sample, based on the availability of recorders and recording

equipment, included one constructed week where recordings were made in three of the markets

(Birmingham, Columbia, and Tampa) and at least three consecutive nights of recordings atvarious times in the other two markets.

The 31 different record dates began fall 2003 in Birmingham (September 8–12, 2003).

Researchers recorded primetime newscasts in that market on that date and again during the

first week of November (the next sweeps period when advertising rates are set based on

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

28 DANIELS AND LOGGINS

rating measures in diaries). Also, from November 10 to 12, 2003, both early evening and

late evening recordings were made in the Columbia market. In spring 2004, newscasts wererecorded in the Tampa–St. Petersburg market from March 3 to 5 and in the Richmond market

from March 31 to April 1. In the interest of consistency across markets, a constructed week

sample was drawn for the May 2004 period where a Monday, Tuesday, Wednesday, Thursday,

and Friday were randomly selected from the 4-week sweeps period and 1 nonsweeps day as a

control. Recordings were made simultaneously in the Birmingham, Columbia, and Tampa–St.Petersburg markets. The six record dates were April 29, May 5, May 7, May 17, May 25,

and June 2. Because of a malfunctioning recorder in the Columbia market on May 17, all

shows from that date were excluded. During the approach of Hurricane Ivan and immediately

following the expansion of one of the older of Birmingham’s two primetime newscast to a

1-hr format, recordings were made on September 12 and 13, 2004. Later that week, recordings

were also made in the Oklahoma City market on September 16, 17, and 18, 2004. Again, thefocus was primarily on primetime newscasts in Oklahoma City. Finally, an additional set of

recordings was made from February 13 to 16, 2005, in the Tampa–St. Petersburg market.

Four stations produced local news programs entirely on their own at least once a day, but

some stations used a different news production strategy. The WB affiliates in Birmingham and

Tampa and the FOX affiliate in Oklahoma City utilized a centralcasting strategy. They usedweathercasts broadcast from a centralized newsroom to stations in the three markets via fiber-

optics technology and produced by AccuWeather, a private weather service. In the Columbia

and Richmond markets, the FOX newscasts were produced under a contract with a competing

station. However, the FOX primetime news program used separately branded weathercasts with

a separate meteorologist.Local newspaper reports and trade magazine market profiles helped determine the top two-

rated stations in each market. Of the 146 weathercasts in this study, 40 aired in Tampa–St.

Petersburg (DMA#13),2 which was the largest of the five markets. In that market, the Media

General–owned NBC affiliate WFLA-TV is the market leader followed by WTVT-TV, a FOX

owned and operated station (Scheiber, 2003). Shortly after the study ended, WTVT extended

its primetime newscast that started at 10 p.m. Eastern time into the 11 p.m. hour. This movelet the station compete head-to-head with the other network affiliates in the late news race

(Squires, 2005).

Two other markets sampled were also top-50 markets. In the Birmingham market (DMA#40),

which includes Anniston and Tuscaloosa, 49 weathercasts were analyzed. The FOX owned and

operated station, WBRC-TV, already produces both a primetime (9 p.m.) and a late newsproduct and has largely dominated for years as the number one station. The number two

station, Allbritton-owned ABC affiliate WBMA-TV, took top honors and the cover of the

RTNDA Communicator magazine for its ratings coup (Cavender, 2005). Slightly smaller than

Birmingham, but covering most of the state of Oklahoma, was the Oklahoma City market

(DMA#45) where 8 weathercasts were analyzed. New York Times–owned KFOR-TV, an NBCaffiliate, has the largest news audience during most newscasts, followed by KOCO-TV, the

ABC affiliate owned by Hearst-Argyle (“Oklahoma TV Newscasts,” 2005).

2Following an increase of 1 million households in the past year, the Tampa–St. Petersburg market moved up to

the 12th largest in the country in the 2005–2006 rankings by Nielsen Media Research (Belcher, 2005).

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

DIFFERENTIATING WEATHER 29

Rounding out the sample were two smaller markets. Virginia’s second largest media mar-

ket, Richmond (DMA#61), covers 32 counties. The Lincoln Financial–owned NBC affiliate,WWBT-TV, is the perennial top-rated station there, followed by CBS affiliate WTVR-TV,

owned by Raycom Media (Romano, 2005). Twelve weathercasts from the Richmond market

were analyzed in this study. Still in the 100 largest designated market areas, but the smallest

of those in this study, Columbia (DMA#83) had 37 weathercasts. The oldest of the stations

in the market, WIS-TV, also broadcasts across the Palmetto State and has remained numberone virtually since its inception. As a former Liberty station, WIS-TV was part of the Raycom

Media acquisition (Werner, 2005). The second-highest rated station is Gannett-owned CBS

affiliate WLTX-TV.

Coding

Two people timed and coded all 146 weathercasts and a separate person verified the predictedtemperatures. The two coders, a graduate student in journalism and a university employee

who worked previously as a television producer, noted the presence or absence of techniques

such as a weather brand (a name for the weathercast), tools or technology (a named weather

instrument), and cross-talk between anchors. “Weather brand” was coded based on whether

a name for the weather segment was audibly or visually (i.e., animation) announced. The

coders also noted the components of the weathercast such as current conditions, satellite map,short-term forecast, and long-term forecast. Finally, the coders noted a predicted high and

predicted weather condition for the next day’s forecast. For comparison, an undergraduate

student transcribed the actual high temperature reported by the National Climactic Data Center

(2003–2004).

Nineteen cases were checked for intercoder reliability on the five primary variables for thisstudy. Where levels of agreement were unacceptable, a second set of 15 cases were cross-

checked using one of the authors of this study as a coder. On those variables, acceptable levels

of consistency between the primary coder (the graduate student) and the author were obtained.

Because the codes were not particularly subjective in nature, accuracy was the primary concern.

For the weather prediction, the coefficient of reliability was .74. For the standard weather fare,the coefficient of reliability was .84. For the crosstalk before and after the weathercast, the co-

efficient of reliability was .86. The announcement of the weather brand coefficient of reliability

was .87 and for announcement of tools and technology the coefficient of reliability was .79.

FINDINGS

Most of the 146 main weathercasts analyzed aired during the primetime or late evening

newscasts. Only 9 of the weathercasts, or 6.2%, were recorded during the early evening news.

The average length of the weathercasts was 3 min 5 sec. The weather predicted was evenly

divided between inclement, noninclement, and in between predictions.Table 1 shows the most common elements of those weathercasts. Radar, satellite maps,

short-term forecasts, and state and regional observations occurred most frequently. They are

considered “Standard Weather Fare.” At the same time, 8 of 10 of the weathercasts included an

extended 5-day forecast, national forecast maps, and current conditions. Only about one fifth

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

30 DANIELS AND LOGGINS

TABLE 1

Most Frequent Weathercast Elements

Weathercast Element Frequency Valid %a

Radarb 139 95.2

Satellite mapsb 136 93.2

Short-term forecastsb 134 91.8

State & regional observationsb 134 91.8

Extended 5-day forecast 124 84.9

National forecast maps 118 80.8

Current conditions 116 79.5

Almanac 65 44.5

Cross-talk after weathercast 64 44.10

Cross-talk before weathercast 57 31.00

Weather video 29 19.90

Live pictures (i.e., towercam) 20 13.70

aValid percentage is out of 100% for each element. bStandard weather fare.

of the weathercasts had weather video or an instance where the weather anchor talked while

video was playing.In other findings, of the 146 weathercasts, 35.6% predicted either scattered showers or

thunderstorms for the next day, whereas 28.7% predicted mostly sunny, partly sunny, or

sunny skies the next day. In between these two extremes, about one fourth (25.7%) of the

weathercasts predicted either partly cloudy skies or scattered clouds for the next day. It is

important to mention none of the weathercasts predicted snow or sleet, probably because ofthe stations’ southern locations. Most of the weathercasts involved traditional introductions

from news anchors. Only 1 in 5 of the weathercasts began with a so-called audio open, where

the weather talent appears after being introduced via sound or animation.

Of the various tools and technologies used for weather product differentiation, radar was,

by far, the most common. Thus, consistent with the literature on “weather wars with radar,”these data provide support for H1: Stations differentiated weathercasts by radar more than other

tools or technology. Examining all the tools and technologies mentioned, only a handful of the

43 weathercasts in that category explicitly mentioned things such as live cameras or satellite

images. FOX owned and operated stations were most likely to make explicit mention of their

radar tools or technology. The FOX weathercasts were 19.9% of the total sample and 44.2%

of stations announcing radar tools or technology.The data also support H2, which suggested that branding was more likely to be found

among the number one and number two rated stations in the market. In fact, 67% of the

branded weathercasts came from one of the top two stations in a market. A chi-square analysis

revealed a significant difference between top stations and other stations use of a weather brand

(�2D 19.86, p < .001). As Table 2 shows, the brand images of these top-rated stations are

fairly similar from market-to-market, with “Storm Team” being the most common brand image

or name. The table also shows the extent to which the top-rated stations extend their brand

image to the naming of tools or technology. Only four (WFLA, WBRC, KOCO, WTVR) of

the stations use both a brand name and a separate brand name for their tools or technology.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

TA

BLE

2

Weath

er

Pro

ducts

of

Mark

et

Leaders

Gro

up

Ow

ner

Len

gth

(min

:sec)

Bra

nd

Ima

ge

(Na

me

of

Wea

therc

ast

)

To

ols

/Tech

no

log

y

(Na

med

On

Air

)

WX

Sea

l

(fo

rW

XC

hie

f)

Tam

pa-

St.

Pet

ersb

urg

,F

L(D

MA

#1

3)

WF

LA

-TV

(NB

C)

Med

iaG

ener

al2

:55

Sto

rmT

eam

8S

torm

Tea

m8

VIP

IRA

MS

WT

VT

-TV

(FO

X)

FO

XB

road

cast

ing

3:2

7S

ky

Tow

erW

eath

erN

ot

men

tio

ned

AM

S

Bir

min

gh

am,

AL

(DM

A#

40

)

WB

RC

-TV

(FO

X)

FO

XB

road

cast

ing

2:5

0F

ox

6S

torm

War

nF

ox

6L

ive

VIP

IR

WB

MA

-TV

(AB

C)

All

bri

tto

n3

:17

No

ne

Th

eS

ky

trac

ker

CB

M

Ok

lah

om

aC

ity,

OK

(DM

A#

45

)

KF

OR

-TV

(NB

C)

New

Yo

rkTim

es

3:2

04

War

nS

torm

Tea

mA

MS

/NW

A

KO

CO

-TV

(AB

C)

Hea

rst-

Arg

yle

3:3

2F

irst

Ale

rtW

eath

erF

irst

Ale

rtS

atel

lite

AM

S

Ric

hm

on

d,

VA

(DM

A#

60

)

WW

BT

-TV

(NB

C)

Jeff

erso

n-P

ilo

t2

:48

Fir

stW

arn

ing

Wea

ther

No

ne

men

tio

ned

AM

S

WT

VR

-TV

(CB

S)

Ray

com

Med

ia3

:19

Sto

rmT

eam

6D

op

ple

rMax

6A

MS

/NW

A

Co

lum

bia

,S

C(D

MA

#8

3)

WIS

-TV

(NB

C)

Lib

erty

Co

rpo

rati

on

3:0

9S

torm

Tea

mN

on

em

enti

on

edA

MS

WL

TX

(CB

S)

Gan

net

tB

road

cast

ing

2:5

6N

on

e1

9L

ive

Do

ub

leD

op

ple

rA

MS

No

te.

AM

SD

Am

eric

anM

eteo

rolo

gic

alS

oci

ety

;C

BM

DC

erti

fied

Bro

adca

stM

eteo

rolo

gis

t;N

WA

DN

atio

nal

Wea

ther

Ass

oci

atio

n.

31

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

32 DANIELS AND LOGGINS

About 60% of branded weathercasts on top-rated stations also ran longer than the mean time

for a weathercast (3 min 5 sec), and nearly all of the meteorologists were credentialed by eitherthe AMS or the National Weather Association.

In the overall sample, only three television stations utilized the “degree-guarantee” promo-

tional strategy. In most cases, the station gave itself a plus or minus 3-degree cushion but

did not show a significantly higher degree of accuracy than other stations. Therefore, the data

do not support H3, which addresses prediction accuracy and guarantees. In response to RQ1,which addressed the overall accuracy of sampled weathercasts, more than half (54.8%) of

the predicted highs were within 3 degrees of the actual high temperature as reported by the

National Weather Service. Twenty-one of the predictions (14.4%) were exactly right. Stations

in the Sinclair Broadcast Group, most of which utilized meteorologists from the AccuWeather

service in a centralized news format, had the highest degree of accuracy of any station group

in the study with 6 perfect predictions and 13 within 3 degrees.RQ2 asked about standard weather fare, or those elements one is most likely to see on a

weathercast regardless of the station. As Table 1 shows, four items emerged in the data that

occur in more than 90% of the 146 weathercasts: radar, satellite maps, short-term forecasts,

and state and regional observations. Another set of three items were very common—occurring

between 80 and 90% of the time: extended or 5-day forecasts, national forecast maps, andcurrent conditions.

In response to RQ3, the data suggest that some strategic planning takes place at the station

group level to brand weathercasts or to emphasize weather as a means of differentiating

products. This was particularly true in the FOX owned and operated stations, which made

the greatest use of branding of any of the stations groups. This was evident on one station’sWeb site—WTVT, in Tampa, boasts of having the largest weather team in the nation. Both

WBRC and WTVT have five staff meteorologists and both now use a branded form of VIPIR

product. Although not branding the weather segments, the Sinclair Broadcast Group had the

most accurate weather predictions presumably, in part, due to the resources of the AccuWeather

service. Other station groups that had multiple stations in this study (Media General, Raycom)

did not appear to have a weather strategy that was as consistent across media markets.

DISCUSSION

As a product, a local station’s weathercast is a tool by which television viewers may differentiate

between choices in a marketplace. The data from this study suggest stations are, in fact, using

the branding strategy to make it easier for viewers to know where to turn. But, as the brandsgo, there isn’t much difference in the name. The analysis of the top two rated stations showed

that meteorologist credentials are no longer a distinguishing factor in that brand. Thus, stations

are dependent on their tools to create vertical product differentiation. However, the data in this

study show that most stations that touted their tools as a reason to buy (watch) their weather

product chose the same tool—radar. Even though a station promotes its accuracy (i.e., degreeguarantee) as a differentiating factor, they are not necessarily any more accurate than other

stations in predicting the next day’s high.

Besides brand names, in the absence of few other product differentiations, perhaps stations

have to look to nonbroadcast delivery methods as a way to make their weather products stand

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

DIFFERENTIATING WEATHER 33

out. As mentioned previously, severe weather brought in 100,000 hits for one Little Rock,

Arkansas, station in 1997 (Galetto, 1997). Also, Randle and Mordock’s (2002) content analysisof home pages showed that television stations offered twice as many weather presentation

devices online as other traditionally offline media. Not among those devices analyzed in the

2002 study was streaming video, which has improved considerably as the number of Web users

with high-speed Web connections has increased. Further research might examine the degree

to which online or other new technologies could serve as a product differentiation factor for alocal station.

Accreditation and focusing on both tools and branding may also be helpful for vertical

product differentiation. A station group that emphasizes weather tools and its weather brand

may experience success in shoring up its market position among the two top stations as

FOX Broadcasting appears to be doing. Considering the many markets where all the chief

meteorologists have AMS accreditation (as shown in Table 2), the AMS has made a wisemove to upgrade its requirements to a new, more difficult Certified Broadcast Meteorologist

seal (Romano, 2005).

It is worth noting that the meteorologists or weather presenters themselves may be a product

differentiation factor. Including the name of the weathercaster (whose name is recognizable

by most viewers) in the video introduction is one way to emphasize what may be a productdifferentiating factor, the person delivering the weather information.

So although a focus on radar or accuracy by themselves may not lead to differentiation,

weather-related Web sites, centralized weather casting to ensure greater accuracy, taking ad-

vantage of increased standards for accreditation, and a focus on branding and tools may help

stations trying to get ahead in the weather wars. We also acknowledge that branding is biggerthan the name used for an item in a television newscast. It can extend to things such as

color combinations, symbols, and other elements that help customers (or viewers) develop an

emotional connection to the product (i.e., weather information) offered.

Despite including five markets ranging in size from the 13th largest to the 83rd largest

designated market area, the convenience sample of this study is only a starting point for a new

round of scholarly research into the weather activities of local broadcast stations. The feasibilityof making recordings over multiple days limits one’s ability to draw an ideal national sample.

However, future research might analyze content over a wider range of stations from some of the

smaller markets (markets larger than 100) as well. This study was based solely on an analysis

of the product without any background knowledge of a station’s weather strategy. A logical

next step would be to contact those producing weather products to understand their intentions,goals, and strategies when producing the weathercasts. Finally, future research might also

extend this assessment beyond the nightly weathercast to the new media techniques stations

are using to differentiate their weather product as well as assess the experience of the individual

weathercasters in a particular market.

REFERENCES

Atwater, T. (1984). Product differentiation in local TV news. Journalism Quarterly, 61, 757–762.

Bae, H.-S. (1999). Product differentiation in cable programming: The case in the cable national all-news networks.

Journal of Media Economics, 12, 265–278.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

34 DANIELS AND LOGGINS

Bae, H.-S. (2000). Product differentiation in national TV newscasts: A comparison of the cable all-news networks and

the broadcast networks. Journal of Broadcasting & Electronic Media, 44(1), 62–77.

Belcher, W. (2005, August 26). Tampa TV market climbs list. Tampa Tribune, p. 1.

Bellamy, R. V., & Traudt, P. J. (2000). Television branding as promotion. In S. T. Eastman (Ed.), Research in media

promotion (pp. 127–159). Mahwah, NJ: Erlbaum.

Bowser, A. (1997, October 27). Weather fronts local news: More viewers say weather is top reason for watching news.

Broadcasting & Cable, 127, 60–64.

Cavender, M. (2005, March). Moving on up. RTNDA Communicator, 59, 22–28.

Chan-Olmsted, S. M., & Kim, Y. (2001). Perceptions of branding among television station managers: An exploratory

analysis. Journal of Broadcasting & Electronic Media, 45(1), 79–91.

Colavecchio-Van Sickler, S. (2004, September 11). Tampa Bay’s weather men. St. Petersburg Times, p. 7A.

Davie, W. H., Auter, P. J., & Dinu, L. F. (2006). Identifying the goals of weather instruction: Toward a model approach

for broadcast meteorology. Journalism and Mass Communication Educator, 61, 149–164.

Dickson, G. (2006, May 22). Coverage goes hyper-local. Broadcasting and Cable, 136(21), 22.

Earl, R. A., & Pasternack, S. (1991). Television weathercasts and their role in geographic education. Journal of

Geography, 90, 113–117.

Galetto, M. (1997). TV weather beat is heating up all over. Electronic Media, 16, 34–36.

Gantz, W. (1982). Redundancy and accuracy of television station weather reports. Journalism Quarterly, 59, 440–446.

Higgins, T. (2005, November 13). Towns hit by tornadoes wake up to debris, disaster. The Des Moines Register, p. 1A.

Hyatt, D., Riley, K., & Sederstrom, N. (1978). Recall of television weather reports. Journalism Quarterly, 55, 306–310.

Ireland, N. J. (1987). Product differentiation and non-price competition. Oxford, UK: Blackwell.

Malone, M. (2007, May 18). Is your weatherman off the mark? Broadcasting & Cable, 137, 10–11.

McDowell, W. S. (2006). Issues in marketing and branding. In A. B. Albarran, S. M. Chan-Olmsted, & M. Wirth

(Eds.), Handbook of media management and economics (pp. 229–273). Mahwah, NJ: Erlbaum.

Meister, M. (2001). Meteorology and the rhetoric of nature’s cultural display. Quarterly Journal of Speech, 87, 415–428.

Murrie, M. (2000, August). Weather wares. RTNDA Communicator, 54, 59.

Murrie, M. (2002, February). Rain, sleet, snow?: New weather products eliminate the guesswork. RTNDA Communi-

cator, 56, 38.

National Climactic Data Center. (2003–2004). Local climatological data reports. Asheville, NC: National Oceanic

and Atmospheric Administration. Retrieved October 2005, from the Edited Local Climatological Data database:

http://ols.nndc.noaa.gov/sub_edlcdsubtype.html

Neal, R., & Hupp, S. (2005, November 8). 20 people still missing. The Indianapolis Star, p. A01.

Neuman, W. R. (1976). Patterns of recall among television news viewers. The Public Opinion Quarterly, 40(1),

115–123.

Oklahoma TV newscasts on KOCO, KFOR take Nielsen titles. (2005, March 4). The Journal Record. Retrieved

November 5, 2005, from http://findarticles.com/p/articles/mi_qn4182/is_20050304/ai_n11849221

O’Malley, S. (1999, December). Weather watch: If there’s one thing viewers tune in for, it’s weather. RTNDA

Communicator, 53, 30–34.

Papper, B. (2002, November). Winning with weather. RTNDA Communicator, 56, 20–25.

Piotrowski, C., & Armstrong, T. R. (1998). Mass media preferences in disaster: A study of Hurricane Danny. Social

Behavior and Personality, 26, 341–345.

Porter, M. E. (1980). Competitive strategy: Techniques for analyzing industries and competitors. New York: The Free

Press.

Potter, S. (2004, July/August). Rating the weather. Weatherwise, 57, 16.

Randle, Q., & Mordock, J. (2002). How radio is adapting weather to the web: A study of weather strategies in local

news/talk radio, television and newspaper home pages. Journal of Radio Studies, 9, 247–258.

Roberts, C. L., & Dickson, S. H. (1984). Assessing quality in local TV news. Journalism Quarterly, 61, 392–398.

Romano, A. (2005, August 29). Capital gains in Virginia. Broadcasting & Cable, 135, 13.

Salsberg, B. (2003, November). Is your station ready for the weather wars? Good weather coverage is honed long

before the big storm hits. RTNDA Communicator, 57, 14–15.

Scheiber, D. (2003, December 16). WTVT No. 1 in key TV viewer group. St. Petersburg Times, p. 2B.

Schneider, M. B. (2005, November 8). Officials: Alerts meant to be heard outdoors. The Indianapolis Star, p. A01.

Schoettle, A. (2005, March 7). TV weather war becoming a race for arms: Local TV ratings, advertising dollars at

stake. Indianapolis Business Journal, 25, 8.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014

DIFFERENTIATING WEATHER 35

Siebert, J. D. (2000). Effects of improved television weather graphics and visual/verbal redundancy on viewer

comprehension. Unpublished doctoral dissertation, The University of New Mexico, Albuquerque.

Squires, C. (2005, November 3). WTVT developing new 11 p.m. newscast. St. Petersburg Times, p. 2B.

Sturken, M. (2001). Desiring the weather: El Nino, the Media and California identity. Public Culture, 13, 161–189.

Tan, A. K. O. (1976). Public media use and preference for obtaining weather information. Journalism Quarterly, 54,

694–705.

Walker, V. (2006). Market size and weathercaster credentials. Feedback, 47(2), 34–40.

Werner, B. (2005, November 2). Raycom plans to sell WACH, Columbia’s FOX affiliate. The State, p. B6.

Wilkins, L. (1984, August). Media coverage of a blizzard: Is the message helplessness? Paper presented at the

Association for Education in Journalism and Mass Communication Conference, Gainesville, FL.

Wulfemeyer, K. T. (1982). A content analysis of local television newscasts: Answering the critics. Journal of Broad-

casting, 20, 481–486.

Wulfemeyer, K. T. (1983). The interests and preferences of audience for local television news. Journalism Quarterly,

60, 323–328.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Car

olin

a] a

t 14:

28 0

8 O

ctob

er 2

014